Dear all, I am doing estimation of parameters of the cox model, and both partial likelihood and full likelihood method have to be used. I knew the coxph can handle with the partial likelihood. But when I assume the baseline hazard function arbitrarily and use full likelihood to do the estimation, I have no idea how to do it with R. Could you please give me any idea whether there is any method of using full likelihood to estimate the parameters of cox model and the baseline function, and how to do it? By the way, I am considering only time dependent covariates in the cox model. Thank you very much! Koshihaku [[alternative HTML version deleted]]